Principal AI Engineer

Recruits Lab exists to connect exceptional life sciences and technology talent with organizations that need them—through relationships, not volume. As Principal AI Engineer you will turn that mission into production systems: models and pipelines that find passive specialists, score fit against real hiring outcomes, and give our recruiters an unfair advantage in markets where the best people never apply.

Performance Objectives

• Ship, within 90 days, a production matching engine that ranks passive life-sciences and tech candidates against live searches using structured outcomes (quota, tenure, named-account signal), not keyword overlap.
• Cut time-to-calibrated-shortlist by 40% in two core practices (life sciences commercial and AI/ML technical) by automating sourcing maps, outreach ranking, and due-diligence prompts recruiters actually use.
• Stand up an evaluation loop that ties model decisions to placement quality—90-day retention, ramp, and client NPS—and retrain quarterly on those labels.
• Own end-to-end reliability of LLM and retrieval systems used in live searches: latency, cost, hallucination rate, and auditability for client-facing work.
• Partner with recruiters to productize two high-leverage workflows (territory mapping and technical-bar screening) so they become default process, not side experiments.
• Define and enforce data contracts, privacy, and bias checks for candidate and client data so we can scale AI without breaking trust.

Environment & Resources

You report to the founding team in a 11–50 person, early-stage firm based in Fort Lauderdale, working remotely-first with recruiters who live in the territories they cover. You will have latitude on stack (Python, modern LLM/RAG tooling, vector search, light MLOps), access to real search data and recruiter feedback, and a mandate to ship rather than prototype. Budget follows proven lift in placement speed and quality.

Essential Qualifications

• Proven ownership of production AI/ML systems that changed a business metric (matching, ranking, retrieval, or agentic workflows)—not research-only or demo work.
• Hands-on depth in LLMs, embeddings, evaluation, and cost/latency tradeoffs; you have shipped and operated, not just prompted.
• Ability to translate messy domain knowledge (quota, buyer types, passive talent maps) into features, labels, and guardrails recruiters will trust.
• Comfort working with small teams: you write the design, the code, the eval, and the runbook.
• Track record of ethical handling of people data and clear communication with non-engineers.

Role Selling Points

• Direct line from model to placement: you will see whether a ranking change filled a histology sales seat in Tri-State or a Principal ML role in Boston.
• Domain that compounds—life sciences and specialized tech hiring is relationship-dense, high-stakes, and poorly served by generic ATS AI.
• Early-stage ownership: architecture, priorities, and standards are still yours to set.
• Mission you can defend: candidate-centric staffing, not spray-and-pray volume.

If you want to build AI that actually moves hard-to-fill searches, apply with a short note on a production system you owned and the metric it moved. We hire the same way we recruit: targeted, calibrated, and serious about outcomes.